基于高密度热辅助交替磁记录技术的新型读写系统研究

Magnetic recording system is still dominant at the storage device market due to its high storage capacity/price ratio. However, the perpendicular magnetic recording system has already approached to 1 Tb/in.^2 limit owing to the superparamagnetic effect. In contrast, the digital information produced in the world has kept increasing by 40% per year. Hence it is urgent to develop the next generation of ultra-high density magnetic recording system to satisfy the need for mass storage. This project plans to develop a novel ultra-high density (10 Tb/in.^2 and above) magnetic recording system, and the corresponding technology challenges are how to successfully write information at such a high density and recover the recorded information in the extremely low signal-to-noise environment. Hence, we plan to investigate the heat-assisted interlaced magnetic recording technology and design novel writing system for the information writing process. The previously written tracks with high laser temperature are trimmed by the latterly written separating tracks with relatively lower laser temperature on both sides and the corresponding transition curvature and adjacent track erasure effects are reduced. Hence the high density information writing process is realized. Additionally, the novel composite media will be studied to mitigate the thermal fluctuation noise that hampers the realization of heat assisted hard disk drive product. Finally, to mitigate the two dimensional noises, interference and nonlinear distortion within the readback signal, the read head array is implemented to simultaneously scan multiple tracks, and the multi-track nonlinear signal processing technologies are investigated, and this decreases the bit error rate and improves the data transfer rate significantly.

磁盘存储系统凭借极高的存储容量造价比在存储器市场内处于主导地位，但由于超顺磁效应的限制，垂直磁记录系统的存储密度已经接近1Tb/in.^2的极限。而目前全世界的数字信息保持着40%的年增速，因此急需研究下一代超高密度磁记录系统以满足海量信息存储的需求。本项目计划研究超高密度(10Tb/in.^2即以上)磁记录系统，其难点是如何实现超高密度下的信息写入，并在极低信噪比环境下恢复记录信息。因此本项目研究了热辅助交替磁记录技术并设计了新型写系统，利用后续低温写入的不相邻磁道双侧剪裁先前高温写入的磁道，减小了信息位弯曲和磁道间擦除效应，从而实现了高密度信息写入。另外研究了新型层合存储介质以减少目前阻碍热辅助磁记录技术产品化的热扰动噪声。最后为了消除读取信号中的二维噪声，窜扰和非线性失真，提出利用读磁头阵列同时读取多条磁道，并研究了多磁道非线性信号检测技术，减小了误码率并提高了数据传输速率。

磁盘存储系统凭借极高的存储容量造价比在存储器市场内处于主导地位，但垂直磁记录系统由于超顺磁效应的限制，存储密度已经接近1Tb/in.^2的极限。而目前全世界的数字信息保持着40%的年增速，因此急需研究下一代超高密度磁记录系统以满足海量信息存储的需求。本项目针对在下一代热辅助交替磁记录(Heat assisted interlaced magnetic recording，HIMR)系统里如何实现超高密度的信息写入，极低信噪比时的准确记录信息恢复进行了一系列的研究探索。本项目首先建立了基于微磁学的简化HIMR读写信道模型，既能刻画近场换能器圆形热光斑引起的信息位弯曲效应以及居里温度分布引起的过渡区噪声，又能满足信号处理需求在短时间内产生海量数据。同时本项目研究了一系列新型读写系统：一方面设计了分离式写磁头以沿跨磁道方向产生非均匀写磁场，从而产生了较笔直的信息位，减小了回读信号的非线性失真；另一方面为了避免常规巨磁电阻(Giant magnetoresistance, GMR)读磁头灵敏度低和滞回严重的缺陷，本项目提出了一种磁力双驱动的巨磁阻抗(Giant magnetoimpedance, GMI)读磁头，通过电场激励压电层所产生的应力和激励线圈所产生的交流磁场协同调控软磁层的磁化强度，从而显著提高了回读信号的信噪比。此外本项目研究了由超顺磁层和存储层构成的层合存储介质，实现了高温下的信息写入过程与长期存储过程相分离，减小了由居里温度分布引起的热扰动噪声。最后为了缓解HIMR系统的二维码间窜扰，介质噪声，信息位弯曲效应对回读信号的非线性失真和衰减，本项目研究了一系列机器学习型多磁道检测方法，包括二维迭代软判决信息辅助型神经网络均衡算法、基于二维软过渡信息辅助型多任务神经网络的多磁道检测方法、联合消除磁道间窜扰和介质噪声的多磁道探测方案、基于改进注意力U-Net的噪声白化多磁道检测算法, 明显降低了多磁道检测过程中的误码率。

内容获取失败，请点击重试